More python 2.7 fun: deprecation of PyCObject API

[Jeff Spaleta]

On Fri, Aug 13, 2010 at 10:20 AM, David Malcolm <dmalcolm at redhat.com> wrote:
> Personally, I'm leaning towards option (a) above (the "don't override
> warnings" option): closing the various as WONTFIX, and adding a section
> to the release notes, whilst working towards fixing this in Fedora 15.
> Affected applications should be patched in Fedora 14 to avoid touching
> the relevant warning setting, and we'll fix the root cause in Fedora 15.
>
> Thoughts?
> Dave

I'll have to read up closely on my upstream numpy and scipy are doing
in their development branches with regard to this.  I think for F14
I'm going to have to live with using the CObject API for these. I
think the next major version of numpy is going to have this mostly
sorted but i'm not keen on pushing it into F14..not unless I have to.

A guarantee you there is a crapload of homebrew academic/science code
out there that uses the CObject API in a numpy oriented workflow.  You
are definitely going to need a release note blurb with a reference
with more detail information on how to work around the problem for
people who run into it with their home brew code.

-jef"not cool"spaleta